Tweet
Houston we have a problem. Thought I'd load your wg as a step file into the mesher as is......Gmesh isn't happy! I haven't looked at your model much but I wonder if there are some relations or constraints missing that tie the mounting structure to the main wg body.
-
-
Ra808> I did two conical guides back at post #323. Only the mouth was rounded over since there is no plane wave at the throat that needs to be bent. The results weren't terribly impressive but I wonder if at that smaller size the mouth roundover is so large it's not really a conical? Perhaps I'll try one at 8".
Nate> I've had an issue (bug?) where when I "extrude" that circular mounting flange only one side gets created. Never figured out why it happened, but in that model it didn't have an issue so I'm surprised to see it show up in your imported model. Other than that, it looks about right, other than not appearing to be 3D.Comment
-
My thoughts are the same regarding the conical horn.... At these small sizes they are basically all round-over. Even at larger sizes where the profile is in use they really don't measure as well as I would hope with some weird things happening off axis. My plan is to investigate alternatives.... No real math involved. Just design and simulate but I'm dealing with compression drivers. augerpro - I'll look at the model more closely this weekend. It is in 3d..... That was a view that showed the error best. Have you thought about how you want to model these things? My thought is that modeling with the wg in an enclosure would be best and would give the best impression of how it will work. Or we could do ib although that will change the directivity where the wg is losing control. I'd guess it'll take less pc horsepower than having the solver process the geometry on the back of the wg which isn't needed anyways.Comment
-
I agree on the "conical section" in a tweeter waveguide.
While re-reading Rod Elliott's article on waveguides, I made some notes.
Rod based his waveguides on Keele's mathematics for Constant Directivity horns (using Keele's Constant).
Even though he clearly distinguishes between compression drivers vs. dome tweeters, stresses the importance of a roundover at the mouth and the termination at the throat, he may not have been fully aware of the throat area as potential cause for interferences affecting performance, especially at high frequencies. This quote is from his website:
"The physics of the situation show that in a duct that has taper, the wave front must be at right angles to the surface and thus cannot be flat, if the taper is very gradual and the wavelength long however, as at low frequencies, the "plane wave" assumption is adequate for practical purposes. It rapidly becomes very inaccurate at higher frequencies, rendering modelling programs of little use for modelling mid and high frequency horns.
In the context of waveguides this means that only a parallel sided duct is in fact a true "waveguide" for a plane wave, and that a flaring duct in order to be called a waveguide has to propagate parallel curved wavefronts. There are only two shapes that satisfy this exactly, the cone for spherical waves, and the spherical sector for cylindrical waves.
All of this might seem a bit academic but in fact it is very relevant to sound quality, the reason for this is most likely the aforementioned multimodal propagation."
The more I think and read about it, the more I am convinced the key to optimal waveguide performance is at the throat, similar to what Geddes found out for compression drivers. Especially for frequencies with wavelengths smaller than the dome (+ surround ?) surface, where the waveguide no longer controls propagation of the wavefronts.
This could also explain - at least to some degree - performance differences between dome and ring radiator tweeters (vifa XT25 etc.).
For a shallower waveguide we may as well connect two curves with different radii, 1 at the throat and another - probably a larger radius - that constitutes the mouth. At, or between the connection points of these 2, a small conical section could optionally be inserted.
Nate, have you tried the STL instead of the .f3d file?
I'll see what the new features and enhancements for imported mesh models in SW 2018 can do. At the moment, I am not in the vicinity of my workstations, so I won't be able to check until next week.Comment
-
From Post #394:
"The physics of the situation show that in a duct that has taper, the wave front must be at right angles to the surface and thus cannot be flat, if the taper is very gradual and the wavelength long however, as at low frequencies, the "plane wave" assumption is adequate for practical purposes. It rapidly becomes very inaccurate at higher frequencies, rendering modelling programs of little use for modelling mid and high frequency horns."
The above paragraph is interesting, particularly the last sentence. It seems an appropriate warning regarding an over reliance on theory in WG design.
Regarding the fit of the WG to the tweeter at the surround, the geometry and construction of the surround can require significant compromise, the ideal, even if assumed to be a specific shape is frequently not practical for a given tweeter, and, as I recall there have been several designs that leave the face plate in place on the tweeter, requiring reconciling an ideal throat configuration to a practical transition and mounting scheme.
I frequently wonder if critical concerns over WG shape reflect a greater concern with theory and testing results, than with actual heard effect. Based upon my experience listening to a variety of tweeters and midrange drivers in various shaped WGs, I suspect that while test results are theoretically assumed to predict certain beneficial effects upon the sound we might hear, these predictions may be less relevant than one might expect, and that the reality is that WG contour is less critical than one might think. There probably are no ideal contours and throat configurations. WGs, like any other part of an audio system require compromises, and at their best can be designed and produced to enhance an individual's listening experience, rather than meeting theoretical design objectives.Comment
-
Ro808 - there was something funky with augerpro's model. I took the liberty to make an adjustment to the mounting plate section and the mesh ran just fine. I'll see if I can get a sim running this weekend.Last edited by natehansen66; 12-02-2017, 11:20 PM.Comment
-
Do you have detailed measurements of the Satori? Height and diameter of dome, height/dia of suspension etc? This thread has me interested in domes for a wg. The factory response and eff of that Satori look promising for a larger wg.Originally posted by augerpro View PostComment
-
Originally posted by natehansen66 View Post
That's good! So you have a CAD model now?
I suppose this isn't sufficient:
http://www.sbacoustics.com/index.php...tori-tw29rn-b/
Otherwise, I'll send a request to SB-Acoustics for detailed data. They may even have a lumped element or baseline driver model, but whether they are willing to give these.... .
By the way, I came across your posts on diyaudio regarding ABEC and the comments from GAGA.
I am a member of that German forum and found some threads on (modeling) horns and waveguides, so I'll snoop around and see if there's some useful input.Last edited by Ro808; 12-03-2017, 04:10 PM.Comment
-
3D printing is gaining popularity. This rectangular OSWG baffle is from a Slovakian forum :
Comment
-
Nate> what did you change? I don't remember what I've done in the past except tweaking a dimension by .001" and then it would produce both flanges properly. Looking at my guide for the TW29 I would approximate the dome height as .09" and the diameter of the surround as 1.68"Comment
-
Looks rough.Originally posted by Ro808 View Post
I wonder how long it required for printing, and what will be used for drivers. No doubt 3D printing will improve and print times be reduced, and perhaps someday wood cell printing will be available. Until then, I think I continue to mill and turn wood baffles.
Comment
-
What's the deal with the PE forum? Fine from tapatalk but on my PC it's super slow and puts weird characters in my posts when I hit enter.
Cool! I've been following a few threads from that forum. I never thought about contacting the manufacturer - let us know what you find please.Originally posted by Ro808 View Post
I do have a CAD model, got the thing meshed and into ABEC but I'm a ways off from solving the model
 Comment
-
Originally posted by augerpro View Post
Thanks for the measurements. The way your model looked in Gmesh it was like your circular baffle mount ring was it's own body without any relations tying it to the main wg. I redrew the ring with your elliptical mouth shape and a circle then extruded it. That worked fine in Gmesh.
Here's how it looks in ABEC with a baffle. I drew a simple dome for the diaphragm. I'm sure I'll need a finer mesh once I figure out how to get the thing to solve. More work to be done!
​
Comment
-
Those flanges probably are seen as different bodies, recently I was told these are referred to as "shells" within the model.Comment
-
Makes sense. I'm used to solidworks. Fusion does things a bit differently.Comment
Comment